uPortal IRC Logs-2008-04-01

Owned by Apereo Infrastructure
Last updated: Apr 01, 2008Version comment
16 min read

[09:32:21 EDT(-0400)] * anastasiac (n=team@142.150.154.149) has joined ##uportal
[09:50:27 EDT(-0400)] * theclown (n=theclown@142.150.154.101) has joined ##uportal
[10:02:45 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1279543721.dsl.bell.ca) has joined ##uportal
[10:10:20 EDT(-0400)] * athena7 (n=athena7@lumina.its.yale.edu) has joined ##uportal
[10:18:31 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:24:36 EDT(-0400)] <athena7> good morning!
[10:24:48 EDT(-0400)] <EricDalquist> morning (smile)
[10:24:58 EDT(-0400)] <EricDalquist> oh athena7 I had a question about the default CAS config
[10:25:02 EDT(-0400)] <athena7> yes
[10:25:15 EDT(-0400)] <EricDalquist> it looks like the cas proxy callback is causing stack traces in system.out and the cas log
[10:25:29 EDT(-0400)] <athena7> hmm
[10:25:47 EDT(-0400)] <athena7> trying to think of when the cas proxy callback actually would be getting called
[10:25:59 EDT(-0400)] <EricDalquist> it should be easily reproducible in the RC3 quickstart
[10:26:01 EDT(-0400)] <EricDalquist> just auth
[10:26:07 EDT(-0400)] <EricDalquist> then look at tomcat/cas.log
[10:28:18 EDT(-0400)] <athena7> yeah
[10:28:22 EDT(-0400)] <athena7> i'll take a look at it
[10:28:25 EDT(-0400)] <athena7> do we have a jira for it?
[10:28:39 EDT(-0400)] <EricDalquist> no, I'll create one next time I start up uportal and get the stack
[10:28:51 EDT(-0400)] <athena7> ok, feel free to assign it to me
[10:28:54 EDT(-0400)] <athena7> that way i won't forget it
[10:28:59 EDT(-0400)] <athena7> i'm a little scattered at the moment
[10:29:02 EDT(-0400)] <EricDalquist> ok (smile)
[10:33:05 EDT(-0400)] <athena7> but our printers are successfully rotating through status messages like "OUT OF LETTERS" and "REPLACE FLUX CAPACITOR"
[10:33:16 EDT(-0400)] <athena7> so today will be a success regardless, i figure
[10:33:23 EDT(-0400)] <EricDalquist> lol
[10:33:24 EDT(-0400)] <EricDalquist> nice
[10:33:32 EDT(-0400)] <athena7> no one seems to have noticed
[10:33:59 EDT(-0400)] <athena7> despite the fact that susan and howard had a long conversation staring at the printer and then replacing one of the ink cartridges (smile)
[10:34:02 EDT(-0400)] <athena7> oh wel!
[10:44:54 EDT(-0400)] * michelled (n=team@142.150.154.199) has joined ##uportal
[11:23:36 EDT(-0400)] <EricDalquist> hrm, athena7 you've worked with Drew's export scripts right?
[11:25:44 EDT(-0400)] <athena7> yeah
[11:26:12 EDT(-0400)] <EricDalquist> I just found an edge case with the layout exports that is causing a channel to get dropped from the export
[11:26:22 EDT(-0400)] <athena7> what's the case?
[11:26:24 EDT(-0400)] <EricDalquist> and I'm not quite sure what to do about it
[11:26:41 EDT(-0400)] <EricDalquist> so in SVN right now the guest.layout has two channels in column 1
[11:26:52 EDT(-0400)] <EricDalquist> the first channel fname doesn't exist
[11:26:54 EDT(-0400)] <EricDalquist> the second does
[11:27:02 EDT(-0400)] <athena7> oh, hm
[11:27:04 EDT(-0400)] <EricDalquist> the layout imports and renders just fine
[11:27:05 EDT(-0400)] <athena7> what's the fname?
[11:27:15 EDT(-0400)] <athena7> that's got to be something i messed up on
[11:27:18 EDT(-0400)] <EricDalquist> uportal-news
[11:27:33 EDT(-0400)] <EricDalquist> the news channel is actually uportal-developers-reference
[11:27:36 EDT(-0400)] <EricDalquist> but that's easy to fix
[11:27:40 EDT(-0400)] <athena7> ok
[11:27:48 EDT(-0400)] <EricDalquist> the problem is the layout export handling the layout structure
[11:28:01 EDT(-0400)] <athena7> is this on the news tab?
[11:28:07 EDT(-0400)] <EricDalquist> \yeah
[11:28:11 EDT(-0400)] <EricDalquist> so
[11:28:16 EDT(-0400)] <athena7> so what does the export produce?
[11:28:22 EDT(-0400)] <EricDalquist> no channels in column1
[11:28:23 EDT(-0400)] <EricDalquist> because
[11:28:34 EDT(-0400)] <EricDalquist> where the uportal-news channel is on import is a null node in the layout
[11:28:47 EDT(-0400)] <athena7> ahh
[11:28:49 EDT(-0400)] <EricDalquist> and the export doesn't know how to handle a leaf node with no channel id
[11:28:59 EDT(-0400)] <athena7> so i guess we need to trap that error
[11:29:23 EDT(-0400)] <EricDalquist> but it needs to understand it so it can follow the next_struct_id to get the next node on that branch
[11:29:27 EDT(-0400)] <athena7> right
[11:29:32 EDT(-0400)] <athena7> that makes sense
[11:29:52 EDT(-0400)] <EricDalquist> I'm looking at the sql in the export layout script and trying to figure out the best way to detect it
[11:30:05 EDT(-0400)] <EricDalquist> the select that gets all folders is: SELECT * FROM up_layout_struct WHERE user_id = 1 AND layout_id = 1 AND chan_id IS NULL;
[11:30:11 EDT(-0400)] <EricDalquist> which returns the null channel node
[11:30:30 EDT(-0400)] <athena7> ohhh
[11:30:37 EDT(-0400)] <EricDalquist> since it has no chan_id
[11:30:55 EDT(-0400)] <EricDalquist> the problem is there is no way to tell what type of node it is with the logic in the script right now
[11:31:03 EDT(-0400)] <EricDalquist> since the script doesn't look at depth
[11:31:04 EDT(-0400)] <athena7> yeah
[11:31:05 EDT(-0400)] <athena7> hm..
[11:31:10 EDT(-0400)] <EricDalquist> it looks at other attributes to determine type
[11:31:16 EDT(-0400)] <athena7> shoudl it be possible for channels to get created with null channel ids?
[11:31:21 EDT(-0400)] <athena7> or does that happen later as an update?
[11:31:42 EDT(-0400)] <EricDalquist> well the issue is at the layout structure level there is no such thing as a channel
[11:31:46 EDT(-0400)] <EricDalquist> just a node with some attributes
[11:31:59 EDT(-0400)] <EricDalquist> the structure XSL assigns meaningful types based on position and depth
[11:32:10 EDT(-0400)] <athena7> yeah
[11:32:49 EDT(-0400)] <athena7> looks like columns generally have a type attribute defined, but don't knwo if that's reliable
[11:32:56 EDT(-0400)] <athena7> regardless, we shoudl be able to trap for that error
[11:32:56 EDT(-0400)] <EricDalquist> yeah
[11:33:07 EDT(-0400)] <athena7> i can take a look at it and talk to drew later on when he gets in
[11:33:35 EDT(-0400)] <EricDalquist> how though ... since it could just as easily be a folder or a column
[11:33:59 EDT(-0400)] <athena7> yeah i guess that's true
[11:34:08 EDT(-0400)] <EricDalquist> and since the export script isn't looking at depth, just attributes :/
[11:34:13 EDT(-0400)] <athena7> anyone know if the type attribute is supposed to be defined for folders?
[11:35:20 EDT(-0400)] <EricDalquist> well in the default database the only nodes with null types are channel nodes
[11:35:25 EDT(-0400)] <EricDalquist> so perhaps we can key off of that
[11:35:45 EDT(-0400)] <athena7> yeah, that's sort of what i'm thinking
[11:36:03 EDT(-0400)] <EricDalquist> eh
[11:36:20 EDT(-0400)] <EricDalquist> in the MyUW QA database we have columns and folders with null types
[11:37:13 EDT(-0400)] <EricDalquist> actually ... maybe not
[11:37:18 EDT(-0400)] <athena7> yeah, that's not surprising
[11:37:36 EDT(-0400)] <athena7> well, we'll figure something out
[11:37:49 EDT(-0400)] <EricDalquist> yeah ... I may try a hand at it
[11:38:10 EDT(-0400)] <EricDalquist> I'm slowly getting closer to having a functional export for 3.0 (smile)
[11:38:18 EDT(-0400)] <EricDalquist> wish I would have tested this back in M5 or RC2 (tongue)
[11:38:34 EDT(-0400)] <athena7> well, you found it now (smile)
[11:39:10 EDT(-0400)] <EricDalquist> yup
[11:39:21 EDT(-0400)] <EricDalquist> and the plus side of this is we're going to reduce our data.xml even further
[11:39:47 EDT(-0400)] <EricDalquist> I think we'll be down to 6 tables that don't have entity files
[11:40:10 EDT(-0400)] <athena7> yeah i saw that
[11:40:12 EDT(-0400)] <athena7> pretty awesome
[11:42:29 EDT(-0400)] * EiNZTEiN (n=einztein@205.241.143.4) has joined ##uportal
[11:43:59 EDT(-0400)] <EiNZTEiN> howdy
[11:44:01 EDT(-0400)] <EiNZTEiN> clear
[11:44:03 EDT(-0400)] <EiNZTEiN> I'm getting a parsing error on web.xml (sad)
[11:44:05 EDT(-0400)] <EiNZTEiN> doesn't make sense though
[11:44:07 EDT(-0400)] <EiNZTEiN> "org.xml.sax.SAXParseException: The element type "web-app" must be terminated by the matching end-tag "</web-app>"."
[11:44:10 EDT(-0400)] <EiNZTEiN> it has the closing tag
[11:44:12 EDT(-0400)] <EiNZTEiN> (sad)
[11:44:24 EDT(-0400)] <EricDalquist> check comments and other unclosed tags
[11:44:52 EDT(-0400)] <EricDalquist> do you have an XML editor on your system? Opening it in one would likely point you to the formatting problem
[11:49:11 EDT(-0400)] <EiNZTEiN> got it actually
[11:49:20 EDT(-0400)] <EiNZTEiN> is there a standard validator for xml?
[11:49:27 EDT(-0400)] <EiNZTEiN> I used one now and it helped find the error
[11:49:49 EDT(-0400)] <EiNZTEiN> being a xhtml/css coder I'm sooo used to validators
[11:49:53 EDT(-0400)] <EiNZTEiN> (smile)
[11:49:54 EDT(-0400)] <EricDalquist> many, I just use the eclipse xml editor as it will flag XML syntax errors
[11:50:13 EDT(-0400)] <EricDalquist> but any editor with XML support should flag invalid XML
[11:56:52 EDT(-0400)] <EricDalquist> man, looking at RDBMUserLayoutStore makes my head hurt
[12:10:14 EDT(-0400)] <EiNZTEiN> haha
[12:10:53 EDT(-0400)] <EiNZTEiN> another thing: I changed by jdbc3 jar to a newer version
[12:10:59 EDT(-0400)] <EiNZTEiN> do I have to update it somewhere?
[12:11:07 EDT(-0400)] <EricDalquist> which version of uPortal?
[12:11:12 EDT(-0400)] <EiNZTEiN> 2.6 (sad)
[12:12:05 EDT(-0400)] <EricDalquist> so if I remember correctly you put it in lib/compile
[12:12:10 EDT(-0400)] <EricDalquist> and update build.properties
[12:13:18 EDT(-0400)] <EiNZTEiN> oh ok
[12:13:24 EDT(-0400)] <EiNZTEiN> i did put it there
[12:13:32 EDT(-0400)] <EiNZTEiN> I'll check build.properties
[12:14:27 EDT(-0400)] <EiNZTEiN> hmm
[12:14:37 EDT(-0400)] <EiNZTEiN> build.properties has no direct file references
[12:14:54 EDT(-0400)] <EricDalquist> just putting it lib/compile may be enough for 2.6
[12:15:09 EDT(-0400)] <EricDalquist> oh
[12:15:22 EDT(-0400)] <EricDalquist> this is 2 ...
[12:15:28 EDT(-0400)] <EiNZTEiN> hehe
[12:15:30 EDT(-0400)] <EricDalquist> so it doesn't go in lib/compile
[12:15:36 EDT(-0400)] <EricDalquist> I think it goes in lib/common
[12:15:52 EDT(-0400)] <EricDalquist> and you'll need to go into TOMCAT_HOME/common/lib and remove the old version of the jar
[12:15:54 EDT(-0400)] <EiNZTEiN> lib/container-common
[12:15:58 EDT(-0400)] <EricDalquist> yeah
[12:16:08 EDT(-0400)] <EiNZTEiN> but there I have the old version (tongue)
[12:16:11 EDT(-0400)] <EricDalquist> uPortal can't easily delete files from that directory since there are a bunch of tomcat jars in there too
[12:16:15 EDT(-0400)] <EricDalquist> yeah
[12:16:20 EDT(-0400)] <EricDalquist> so delete the old version
[12:16:22 EDT(-0400)] <EricDalquist> and put the new one in
[12:16:25 EDT(-0400)] <EiNZTEiN> the old version of the jdbc jar
[12:16:44 EDT(-0400)] <EiNZTEiN> but it's complaining it can't find the old version
[12:16:45 EDT(-0400)] <EiNZTEiN> :o
[12:17:02 EDT(-0400)] <EricDalquist> what is complaining?
[12:17:04 EDT(-0400)] <EricDalquist> and what is the message?
[12:17:11 EDT(-0400)] <EiNZTEiN> I'll tell ya in a sec
[12:17:16 EDT(-0400)] <EricDalquist> heh
[12:17:21 EDT(-0400)] <EiNZTEiN> (smile)
[12:18:13 EDT(-0400)] <EiNZTEiN> don't forgetting I'm using a very customized uPortal for the Portfolios channel, by Pete Boysen
[12:18:14 EDT(-0400)] <EiNZTEiN> (tongue)
[12:18:48 EDT(-0400)] <EricDalquist> yup, so remember that we may not be able to help much because of that (tongue)
[12:20:27 EDT(-0400)] <EiNZTEiN> ya I know (wink)
[12:22:19 EDT(-0400)] <EiNZTEiN> I hate using things I can't comprehend
[12:22:20 EDT(-0400)] <EiNZTEiN> lol
[12:23:13 EDT(-0400)] * awills (n=awills@12.164.136.183) has joined ##uportal
[12:23:23 EDT(-0400)] <awills> hey folks
[12:23:26 EDT(-0400)] <EricDalquist> hey awills
[12:23:34 EDT(-0400)] <EiNZTEiN> howdy awills
[12:23:38 EDT(-0400)] <EricDalquist> great timeing I think I might have a fix for the issue I've been working on (smile)
[12:24:03 EDT(-0400)] <EricDalquist> I found an instance where an incomplete layout was being exported
[12:24:27 EDT(-0400)] <awills> interesting... what kind of circumstances
[12:24:33 EDT(-0400)] <EricDalquist> so the guest.layout in SVN right now has two channels in the first column, the first channel does not exist and the second does
[12:24:41 EDT(-0400)] <EricDalquist> this results in a valid layout document
[12:24:46 EDT(-0400)] <EricDalquist> with a null node in it
[12:25:17 EDT(-0400)] <awills> which doesn't work on import, i imagine
[12:25:38 EDT(-0400)] <EricDalquist> import worked
[12:25:43 EDT(-0400)] <EricDalquist> rendering in the portal worked
[12:26:01 EDT(-0400)] <EricDalquist> export worked, but was missing all children and siblings of the null node
[12:26:16 EDT(-0400)] <EricDalquist> so since this was the first channel of two in a column the second channel wouldn't show up in the export
[12:26:53 EDT(-0400)] <EricDalquist> it works in the portal since the portal builds a tree structure then converts the generic folder/channel tree into something more meaningful based on node position and depth
[12:27:10 EDT(-0400)] <EricDalquist> where as the export script uses node attributes to determine the type of each node
[12:28:02 EDT(-0400)] <EricDalquist> and this node has no attributes, just a struct_id and next_struct_id
[12:28:59 EDT(-0400)] <awills> this node is a channel node that's missing some data?
[12:29:34 EDT(-0400)] <EricDalquist> so, without completely redoing the export script to recursively build a tree (which seems like a lot of work) I think I have a solution: http://uportal.pastebin.com/d64681f26
[12:29:38 EDT(-0400)] <EricDalquist> yeah
[12:29:44 EDT(-0400)] <EricDalquist> but in theory it could be any node in the tree
[12:30:02 EDT(-0400)] <EricDalquist> that only has a struct_id and then possibly a next and/or child id
[12:30:21 EDT(-0400)] <EricDalquist> the fix I just posted only handles the null channel node case
[12:30:48 EDT(-0400)] <EricDalquist> I'm not sure how you'd handle a null folder case without refactoring the export to actually build the tree
[12:31:39 EDT(-0400)] <EiNZTEiN> EricDalquist,
[12:31:41 EDT(-0400)] <EiNZTEiN> WARNING: A docBase /opt/local/share/java/tomcat5/webapps/uPortal inside the host appBase has been specified, and will be ignored
[12:31:41 EDT(-0400)] <EiNZTEiN> Apr 1, 2008 11:31:02 AM org.apache.catalina.core.StandardContext start
[12:31:41 EDT(-0400)] <EiNZTEiN> SEVERE: Error listenerStart
[12:31:43 EDT(-0400)] <EiNZTEiN> ?
[12:31:51 EDT(-0400)] <EiNZTEiN> any ideas?
[12:32:09 EDT(-0400)] <EricDalquist> not from that, are there errors in the portal log file?
[12:32:15 EDT(-0400)] <EiNZTEiN> I'll check
[12:32:16 EDT(-0400)] <EricDalquist> if so paste those via pastebin
[12:32:23 EDT(-0400)] <EiNZTEiN> ya
[12:32:32 EDT(-0400)] <EiNZTEiN> I just pasted those cuz it was 3 lines (tongue) sorryyyy
[12:33:33 EDT(-0400)] <EiNZTEiN> hmm
[12:33:45 EDT(-0400)] <EiNZTEiN> when your run tomcat from eclipse it doesn't log to catalina.out? :o
[12:33:59 EDT(-0400)] <EricDalquist> nope
[12:34:03 EDT(-0400)] <EricDalquist> just to the console
[12:34:19 EDT(-0400)] <EricDalquist> all catalina.out is from is the startup script redirection stdout to the file catalina.out
[12:34:26 EDT(-0400)] <EricDalquist> you need to look for the uPortal log file though
[12:34:35 EDT(-0400)] <EiNZTEiN> so where can I get the logs you're asking from?
[12:34:38 EDT(-0400)] <EiNZTEiN> the portal log
[12:35:01 EDT(-0400)] <EricDalquist> uPortal/properties/Logger.properties defines the portal's logging config
[12:35:08 EDT(-0400)] <EricDalquist> look in there for the file name/location
[12:35:18 EDT(-0400)] <awills> EiNZTEiN I also get that message in tomcat, in all my portals afaik
[12:35:23 EDT(-0400)] <EricDalquist> if there is no path set it will end up in the 'working directory' for tomcat
[12:36:11 EDT(-0400)] <EricDalquist> which may just be TOMCAT_HOME
[12:37:36 EDT(-0400)] <EiNZTEiN> it's inside uPortal root
[12:37:37 EDT(-0400)] <EiNZTEiN> (smile)
[12:37:38 EDT(-0400)] <EiNZTEiN> anyway
[12:37:40 EDT(-0400)] <EiNZTEiN> http://uportal.pastebin.com/d743da5bf
[12:37:58 EDT(-0400)] <EricDalquist> looks fine
[12:38:06 EDT(-0400)] <EricDalquist> like everything started as expected
[12:38:19 EDT(-0400)] <EricDalquist> does it not work when you try browsing to it?
[12:39:15 EDT(-0400)] <EiNZTEiN> nop
[12:39:21 EDT(-0400)] <EricDalquist> what do you see?
[12:39:22 EDT(-0400)] <EiNZTEiN> and I can't deploy it through tomcat's manager
[12:39:24 EDT(-0400)] <EiNZTEiN> 404
[12:39:27 EDT(-0400)] <EricDalquist> in the browser
[12:39:49 EDT(-0400)] <EiNZTEiN> HTTP Status 404 - /uPortal/
[12:39:54 EDT(-0400)] <EricDalquist> and that paste is the entire .log file?
[12:40:04 EDT(-0400)] <EiNZTEiN> for this occurence
[12:40:14 EDT(-0400)] <EiNZTEiN> I'll delete the log and re-run it
[12:40:16 EDT(-0400)] <EiNZTEiN> w8 a sec
[12:40:29 EDT(-0400)] <EricDalquist> according to that log it should be up
[12:40:38 EDT(-0400)] <EricDalquist> awills: any thoughts?
[12:40:46 EDT(-0400)] <EricDalquist> on the layout export?
[12:40:55 EDT(-0400)] <awills> EricDalquist it seems as though this trouble occurs when the chan_fname field is null?
[12:41:05 EDT(-0400)] <EricDalquist> in this case yes
[12:41:09 EDT(-0400)] <awills> on a row that should be a channel?
[12:41:19 EDT(-0400)] <EricDalquist> a layout node was created for a channel but no chan_id was specified
[12:41:33 EDT(-0400)] <EricDalquist> fixing the import script may be something we want to look at too
[12:41:50 EDT(-0400)] <EricDalquist> but the fact that this layout is apparently valid for the portal makes me want to ensure it can be exported correctly
[12:41:51 EDT(-0400)] <awills> and the <null-node> structure, and the SQL query that finds them, is designed to handle this case?
[12:42:27 EDT(-0400)] <awills> your query: SELECT * FROM up_layout_struct WHERE user_id = ? AND layout_id = 1 AND chan_id IS NULL AND CHLD_STRUCT_ID IS NULL
[12:43:25 EDT(-0400)] <awills> suggests to me that you've figured out that the CHLD_STRUCT_ID filed should be null for all channel rows
[12:43:28 EDT(-0400)] <EricDalquist> yeah, that should find leaf nodes with no channel id
[12:43:39 EDT(-0400)] <EricDalquist> yeah
[12:43:55 EDT(-0400)] <EricDalquist> as far as I can tell and remember <channel> nodes can't have children
[12:44:12 EDT(-0400)] <awills> i think we can change the above SQL, the one that handles regular (non-null) channels
[12:44:25 EDT(-0400)] <awills> now it's: SELECT * FROM up_layout_struct WHERE user_id = ? AND layout_id = 1 AND chan_id IS NULL
[12:44:45 EDT(-0400)] <awills> we can make it: SELECT * FROM up_layout_struct WHERE user_id = ? AND layout_id = 1 AND chld_struct_id IS NULL
[12:45:09 EDT(-0400)] <awills> so the null channels will come in w/ the rest of them
[12:45:30 EDT(-0400)] <awills> oops wrong paste
[12:45:44 EDT(-0400)] <EiNZTEiN> http://uportal.pastebin.com/d46701fd1 << this is the error I get from eclipse when trying to run tomcat
[12:45:47 EDT(-0400)] <EiNZTEiN> :/
[12:45:51 EDT(-0400)] <EiNZTEiN> not helpful et all
[12:45:59 EDT(-0400)] <EricDalquist> ok
[12:46:13 EDT(-0400)] <EricDalquist> so when you see that there should be an error message in the portal log file
[12:46:21 EDT(-0400)] <EiNZTEiN> ok
[12:46:32 EDT(-0400)] <awills> eric let me play a bit
[12:46:32 EDT(-0400)] <EiNZTEiN> I got the complete log now (after deleting the huge old file
[12:47:05 EDT(-0400)] <EiNZTEiN> http://uportal.pastebin.com/d74a800f1
[12:47:14 EDT(-0400)] <EricDalquist> awills: ok
[12:48:07 EDT(-0400)] <EricDalquist> and that is the entire .log file? that doesn't seem complete
[12:48:27 EDT(-0400)] <EiNZTEiN> it is :S
[12:49:02 EDT(-0400)] <EiNZTEiN> here is the log in eclipse
[12:49:03 EDT(-0400)] <EiNZTEiN> http://uportal.pastebin.com/d1dca51ab
[12:49:37 EDT(-0400)] <EricDalquist> yeah, there should be an error in the portal log detailing what happened during startup
[12:49:54 EDT(-0400)] <EricDalquist> it looks like it is in the middle of loading the Spring application context then the log just stops
[12:50:02 EDT(-0400)] <EricDalquist> no completion message and no failure message
[12:51:22 EDT(-0400)] <EiNZTEiN> argh
[12:52:13 EDT(-0400)] <EricDalquist> can you pastebin your Logger.properties file?
[12:52:25 EDT(-0400)] <EiNZTEiN> y
[12:52:55 EDT(-0400)] <EiNZTEiN> http://uportal.pastebin.com/d68644207
[12:54:27 EDT(-0400)] <EricDalquist> looks like the default log config ... yeah I have no idea why your portal.log just ends like that
[12:55:02 EDT(-0400)] <EricDalquist> the error in the eclipse console happens when an exception is thrown while tomcat is initializing the webapp
[12:55:16 EDT(-0400)] <awills> EricDalquist – I think there are 2 problems w/ that row on up_layout_struct: (1) it doesn't get picked up by the channel SQL, and (2) it will get picked up by the folder sql
[12:55:46 EDT(-0400)] <EricDalquist> :/
[12:56:04 EDT(-0400)] <EricDalquist> so how hard would it be to have the layout export actually walk the layout tree?
[12:56:13 EDT(-0400)] <awills> (1) occurs b/c there's a join on the up_channel table... we don't need to do a join there, i can reword that query (and xml that uses it) not to do a join
[12:56:24 EDT(-0400)] <EricDalquist> EiNZTEiN: check in tomcat/logs
[12:56:32 EDT(-0400)] <EricDalquist> see if there is a catalina.log file to look at
[12:57:33 EDT(-0400)] <EricDalquist> awills: ok ... is this something you want to work on or should I continue with it?
[12:58:22 EDT(-0400)] <awills> i can take it... shouldn't be too long
[12:58:27 EDT(-0400)] <EricDalquist> ok
[12:58:36 EDT(-0400)] <awills> i also agree the import should at least issue a warning
[12:58:39 EDT(-0400)] <EricDalquist> you can easily reproduce it with the trunk right now
[12:58:53 EDT(-0400)] <EiNZTEiN> EricDalquist, ok
[12:59:59 EDT(-0400)] <awills> yeah i'm looking at the layout document now
[13:00:06 EDT(-0400)] <EiNZTEiN> on tomcat/logs I only have old logs from when I installed tomcat and ran it from the cmd line instead of eclipse
[13:00:10 EDT(-0400)] <awills> column 1 has no channels
[13:00:19 EDT(-0400)] <EiNZTEiN> the last modified date is from 3/26
[13:01:10 EDT(-0400)] <EricDalquist> EiNZTEiN: I'm not sure what to tell you, perhaps try changing the log level in Logger.properties to debug
[13:01:15 EDT(-0400)] <EricDalquist> then deploy
[13:01:27 EDT(-0400)] <EricDalquist> then try starting again and see if there is anything more interesting in the log file
[13:01:38 EDT(-0400)] <EiNZTEiN> ok
[13:08:44 EDT(-0400)] <EiNZTEiN> from INFO to DEBUG?
[13:08:59 EDT(-0400)] <EricDalquist> yup
[13:09:12 EDT(-0400)] <EiNZTEiN> log4j.rootCategory=DEBUG, R < ?
[13:09:42 EDT(-0400)] <EricDalquist> oh and just so people know, there is a wisclist outage
[13:10:10 EDT(-0400)] <EricDalquist> well not completely out but they have a 4+ hour mail queue right now
[13:11:26 EDT(-0400)] <EiNZTEiN> lunch time
[13:11:28 EDT(-0400)] <EiNZTEiN> ttyl (smile)
[13:11:34 EDT(-0400)] <athena7> eek, sad mail server
[13:14:15 EDT(-0400)] <EricDalquist> http://kb.wisc.edu/helpdesk/outage.php?action=5&amp;caseID=080401-0103
[13:15:32 EDT(-0400)] <awills> here you go eric: http://uportal.pastebin.com/d580d0115
[13:16:25 EDT(-0400)] <EricDalquist> great, thanks!
[13:17:17 EDT(-0400)] <awills> item (2) above – null channels appearing as folders – isn't really an issue b/c the XSL won't ever include them since it only accepts channels as children of columns
[13:17:36 EDT(-0400)] <awills> is there a good jira?
[13:18:08 EDT(-0400)] <EricDalquist> not for that specific issue yet
[13:18:48 EDT(-0400)] <awills> i'll make one
[13:19:08 EDT(-0400)] <EricDalquist> thanks!
[13:24:46 EDT(-0400)] <awills> http://www.ja-sig.org/issues/browse/UP-1989
[13:27:16 EDT(-0400)] <awills> all done
[13:28:11 EDT(-0400)] <EricDalquist> great thanks!
[13:36:07 EDT(-0400)] <EricDalquist> lists should be fixed
[13:38:42 EDT(-0400)] <athena7> yea!
[13:38:47 EDT(-0400)] * athena7 cheers
[13:43:59 EDT(-0400)] <EricDalquist> athena7: http://www.ja-sig.org/issues/browse/UP-1990
[13:44:09 EDT(-0400)] <athena7> oh good, thank you
[13:44:13 EDT(-0400)] <athena7> i'll look at it
[13:46:55 EDT(-0400)] <athena7> eric do you do IM?
[13:47:04 EDT(-0400)] <EricDalquist> yup
[14:03:38 EDT(-0400)] <EricDalquist> heh, so this really slow vendor tool that a co-worker has been dealing with
[14:03:48 EDT(-0400)] <EricDalquist> we just ran the explain plan in Oracle on their query
[14:04:15 EDT(-0400)] <EricDalquist> O(N^N)
[14:04:17 EDT(-0400)] <EricDalquist> yay!
[14:04:47 EDT(-0400)] <EricDalquist> I think on our data set it works out to (2,000,000 * 6,000,000) row accesses
[14:07:14 EDT(-0400)] <athena7> omg.
[14:08:27 EDT(-0400)] <EricDalquist> yup
[14:08:32 EDT(-0400)] <EricDalquist> pretty sweet huh
[14:08:40 EDT(-0400)] <athena7> yeah
[14:08:41 EDT(-0400)] <EricDalquist> that is their impl of what is just a depth-first search of a tree
[14:09:06 EDT(-0400)] <athena7> oh my
[14:09:09 EDT(-0400)] <EricDalquist> and a tree that only has ~ 3mil nodes
[14:09:17 EDT(-0400)] <EricDalquist> so big, but not anything crazy
[14:09:53 EDT(-0400)] <athena7> sounds lovely (smile)
[14:10:02 EDT(-0400)] <EricDalquist> I think my CS algorithms proof would have blew a gasket (smile)
[14:10:21 EDT(-0400)] <athena7> i'm sure!
[14:11:27 EDT(-0400)] <EricDalquist> awills: I have a crn question for you
[14:12:03 EDT(-0400)] <EricDalquist> can I set a variable value in node a and access it in node b: <s><a/><b/></s>
[15:34:29 EDT(-0400)] * dstn (n=dstn@unaffiliated/dstn) has joined ##uportal
[15:47:04 EDT(-0400)] <EricDalquist> hey athena7, that skin picker problem, did that ever get fixed?
[15:47:21 EDT(-0400)] <athena7> the fix you checked in worked
[15:47:31 EDT(-0400)] <athena7> and i discovered there there were a couple javascript bugs too
[15:47:34 EDT(-0400)] <athena7> i haven't checked that in yet
[15:47:46 EDT(-0400)] <EricDalquist> I'm not sure I ever checked it in ... I just ran into it again (smile)
[15:47:50 EDT(-0400)] <athena7> it was pulling the cached skin xml rather than updating it
[15:48:03 EDT(-0400)] <athena7> oh, what specfically?
[15:48:15 EDT(-0400)] <EricDalquist> pstmt.executeUpdate(sQuery);
[15:48:15 EDT(-0400)] <athena7> maybe i applied it manually?
[15:48:16 EDT(-0400)] <athena7> hm.
[15:48:22 EDT(-0400)] <athena7> oh ok i must have patched it locally
[15:48:27 EDT(-0400)] <athena7> i can check that in if you'd like
[15:48:27 EDT(-0400)] <EricDalquist> yeah I think I forgot to doc and commit it
[15:48:28 EDT(-0400)] <EricDalquist> oops
[15:48:31 EDT(-0400)] <EricDalquist> I'll go fix that
[15:48:32 EDT(-0400)] <EricDalquist> (smile)
[15:48:33 EDT(-0400)] <athena7> ok
[15:48:39 EDT(-0400)] <athena7> and i'll check in the javascript fix (smile)
[15:48:48 EDT(-0400)] <athena7> oh!
[15:48:50 EDT(-0400)] <athena7> i remember
[15:48:54 EDT(-0400)] <athena7> turns out the universality.xsl
[15:48:59 EDT(-0400)] <athena7> sets the parameter name to SKIN
[15:49:01 EDT(-0400)] <athena7> rather than skin
[15:49:05 EDT(-0400)] <athena7> so it doesn't pick up the value in the db
[15:49:10 EDT(-0400)] <athena7> so i need to check in that fix
[15:49:34 EDT(-0400)] <EricDalquist> ah
[15:51:10 EDT(-0400)] <EricDalquist> so I'm about to commit an updated data.xml and set of entity files
[15:51:22 EDT(-0400)] <EricDalquist> I have noticed a bunch of fnames and channel titles don't correspond to what the channel is
[15:51:39 EDT(-0400)] <EricDalquist> like groups manager on the admin layout is names Tomcat Servlet Examples
[15:54:46 EDT(-0400)] <athena7> yeah
[15:55:06 EDT(-0400)] <athena7> hm, didn't realize it was that bad (smile)
[15:55:10 EDT(-0400)] <athena7> ok i'll be back in a few minutes
[15:55:14 EDT(-0400)] <athena7> running out to grab coffee
[16:07:21 EDT(-0400)] <EiNZTEiN> back
[16:07:25 EDT(-0400)] <EiNZTEiN> (with coffee (tongue))
[16:07:50 EDT(-0400)] <dstn> nice!
[16:07:57 EDT(-0400)] <EiNZTEiN> (and renewed drivers license (tongue))
[16:08:13 EDT(-0400)] <EiNZTEiN> i was driving with it having expired almost a month ago
[16:08:13 EDT(-0400)] <EiNZTEiN> lol
[16:08:17 EDT(-0400)] <EiNZTEiN> yikes
[16:14:55 EDT(-0400)] <EricDalquist> data.xml is down to 429 lines!
[16:17:15 EDT(-0400)] <athena7> yea!! (smile)
[16:17:53 EDT(-0400)] <EricDalquist> committing now ... this is a big changeset
[16:22:24 EDT(-0400)] <EricDalquist> heh still committing ... (smile)
[16:25:51 EDT(-0400)] <levifig> EricDalquist, im uploading the "debug" version of portal.log now (pastebin)
[16:26:06 EDT(-0400)] <EricDalquist> ok
[16:26:12 EDT(-0400)] <levifig> i don't think pastebin will handle (tongue)
[16:26:20 EDT(-0400)] <levifig> it did
[16:26:20 EDT(-0400)] <levifig> (tongue)
[16:26:21 EDT(-0400)] <levifig> http://uportal.pastebin.com/d651f043
[16:27:11 EDT(-0400)] <athena7> hmmm - shouldn't something like "ant crn-import -Ddir=/Users/jdb53/Documents/workspace/uportal/MailPortlet/docs -Dpattern=*" work in up3?
[16:27:35 EDT(-0400)] <EricDalquist> I would guess so
[16:27:37 EDT(-0400)] <EricDalquist> ok
[16:27:42 EDT(-0400)] <EricDalquist> my changes are committed
[16:27:45 EDT(-0400)] <EricDalquist> it will be quite the update
[16:28:51 EDT(-0400)] <EricDalquist> still looks good levifig ... are you sure it is deploying to the /uPortal context and not some other name due to the customizations?
[16:29:25 EDT(-0400)] <levifig> hmm
[16:29:37 EDT(-0400)] <levifig> shoud
[16:29:39 EDT(-0400)] <levifig> should*
[16:29:46 EDT(-0400)] <levifig> I have nothing else on my tomcat :S
[16:29:54 EDT(-0400)] <athena7> awills: are you around?
[16:30:59 EDT(-0400)] <EricDalquist> are you seeing an error athena7?
[16:31:11 EDT(-0400)] <athena7> yeah
[16:31:18 EDT(-0400)] <athena7> works ok when i give it the actual file name
[16:31:25 EDT(-0400)] <athena7> i thought i'd used this recently though so i'm a little confused
[16:31:41 EDT(-0400)] <EricDalquist> oh
[16:31:42 EDT(-0400)] <EricDalquist> .*
[16:31:45 EDT(-0400)] <EricDalquist> not *
[16:31:48 EDT(-0400)] <EricDalquist> I think it is a regex
[16:32:04 EDT(-0400)] <athena7> ohhh
[16:32:17 EDT(-0400)] <athena7> that would totally make sense
[16:32:27 EDT(-0400)] <athena7> excellent, yes, that works (smile)
[16:32:28 EDT(-0400)] <athena7> thanks
[16:32:33 EDT(-0400)] <EricDalquist> yup
[16:32:39 EDT(-0400)] <athena7> ok, great
[16:32:43 EDT(-0400)] <athena7> all is right with the world again
[16:32:51 EDT(-0400)] <athena7> i could have sworn i'd used it before (smile)
[16:34:15 EDT(-0400)] <athena7> eric do you have any thoughts about the yale mods i'd mentioned before to the bookmarks portlet? (using remote xml files as folders)
[16:34:27 EDT(-0400)] <athena7> is that something you'd be interested in having contributed?
[16:34:33 EDT(-0400)] <EricDalquist> sounds interesting, yeah
[16:34:41 EDT(-0400)] <athena7> ok
[16:34:49 EDT(-0400)] <athena7> maybe i could write up a summary of work to date and send it to the portlets list?
[16:35:01 EDT(-0400)] <awills> hey folks, i'm here
[16:35:04 EDT(-0400)] <EricDalquist> that would be good
[16:35:04 EDT(-0400)] <athena7> it's not completely done, and you may have some architectural suggestions, since you know the code a lot better
[16:35:09 EDT(-0400)] <awills> just fighting eclipse
[16:35:14 EDT(-0400)] <athena7> awills: eric sorted out my problem for me, thanks (smile)
[16:35:23 EDT(-0400)] <awills> eric's got it – .*
[16:35:39 EDT(-0400)] <athena7> ok, great - i'd like to get as much yale code into public repositories as possible
[16:35:52 EDT(-0400)] <EricDalquist> http://jasig.princeton.edu/source/changelog/jasigsvn?cs=43469&amp;csize=575
[16:36:05 EDT(-0400)] <EricDalquist> there is the commit that drops all but 6 tables from data.xml
[16:36:13 EDT(-0400)] <EricDalquist> and uses human readable entity file names
[16:36:14 EDT(-0400)] <awills> nice
[16:37:02 EDT(-0400)] <athena7> awesome
[16:37:32 EDT(-0400)] <EricDalquist> looks like it is a near identity transform now
[16:37:42 EDT(-0400)] <EricDalquist> I did an import then an export
[16:37:54 EDT(-0400)] <EricDalquist> diffd the two and only very minor XML element ordering differences
[16:38:06 EDT(-0400)] <awills> that's great
[16:49:16 EDT(-0400)] <EricDalquist> jen fyi I created http://www.ja-sig.org/issues/browse/UP-1991 for the channel entity cleanup
[16:49:25 EDT(-0400)] <athena7> ok, thanks, sounds great
[16:49:42 EDT(-0400)] <EricDalquist> if you'd like I can actually spend a little time on that right now
[16:50:26 EDT(-0400)] <athena7> it would probably be good to clean that up i guess
[16:50:31 EDT(-0400)] <athena7> it's pretty confusing
[16:51:53 EDT(-0400)] <levifig> gotta reboot folks
[16:51:56 EDT(-0400)] <levifig> cya in a bit
[17:02:07 EDT(-0400)] * awills (n=awills@12.164.136.183) has left ##uportal
[17:14:36 EDT(-0400)] * michelled (n=team@142.150.154.199) has left ##uportal
[17:27:14 EDT(-0400)] * levifig (n=einztein@205.241.143.4) has joined ##uportal
[17:48:25 EDT(-0400)] * theclown (n=theclown@142.150.154.101) has left ##uportal
[18:53:47 EDT(-0400)] * athena7 (n=athena7@c-24-2-189-230.hsd1.ct.comcast.net) has joined ##uportal